Skip to content
Other

StreamDiffusion V2 ONNX

video model export for fast inference

2 min readupdated

This repository provides a ready-to-run ONNX export (a format for optimized model execution) of the causal video-to-video model StreamDiffusionV2, built on Wan2.1 1.3B. The bundle includes the DMD DiT components (in fp16 at 256x256 resolution with a KV-cache batch profile of 1-4), the UMT5-XXL text encoder, the TAEHV VAE autoencoder, and a tokenizer. Built with TiXL tools, it is designed for third-party apps and model browsers, letting you run it without installing Python on the target machine.

Key features

  • Based on the causal Wan2.1 1.3B DiT architecture for video generation.
  • Exported to ONNX format (fp16) at 256x256 resolution.
  • Includes the UMT5-XXL text encoder and TAEHV autoencoder.
  • Supports a KV-cache batch profile from 1 to 4.
  • Bundles every component needed to run via TiXL without Python.

What you can do

  • Integrate the ready-to-use model into TiXL-compatible apps or model browsers.
  • Run video generation on target devices without installing Python.

Example

The model ships as pre-built ONNX files ready to export and run via TiXL utilities; no code snippet is included in the model card.

Read moreDetails
License: Apache-2.0 (for the base StreamDiffusionV2 and Wan2.1-T2V-1.3B weights) and MIT (for the TAEHV autoencoder by madebyollin/taehv).
Components: DMD DiT (fp16, 256x256, KV-cache batch profile 1-4), UMT5-XXL, TAEHV VAE, and tokenizer.
Discuss Telegram 0

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 0.0 GB · ready files0.5 GB
  • FP8 / INT8recommendedweights 0.0 GB0.5 GB
  • 4-bitweights 0.0 GB0.5 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · BF16 — from RTX 2060

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # banidu-512/streamdiffusion-v2-onnx
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download banidu-512/streamdiffusion-v2-onnx --local-dir ~/models/streamdiffusion-v2-onnx
    # The model is in ~/models. How to run it is in the model card (link above).
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of streamdiffusion-v2-onnx (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

banidu-512/streamdiffusion-v2-onnx
Author
banidu-512
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments